706 C. E. McCLUNG 



continuous with the quarter ring from which it arose. In other 

 words, it may be stated that each prophase ring is divided into 

 superimposed half rings by the longitudinal division of the con- 

 joined homologous chromosomes, and that, by a plane at right 

 angles to this, and passing through the points of endwise union 

 of the homologous chromosomes, each half ring is divided into 

 quarter rings. There is, then, no difference between a rod, V 

 or ring except that in the latter the free ends are joined as at the 

 opposite synaptic ends. Uniformity of construction is main- 

 tained. Such an interpretation of annular tetrads is given by 

 Sutton ('02, figs. 5 to 7) ; Baumgartner ('04) ;Pinney ('08, figs. 18 a, 

 d, e, f, h, 21); Davis ('08, figs. 59 to 61, 85, 170 to 178); Rob- 

 ertson ('08, figs. 26, 29); Buchner ('09, fig. 41); Granata ('10, 

 figs. 27 to 29, text fig. 1 g, h, i, 1, m, n, o, p). 



A modification of this method of ring production consists in 

 the simple approximation of the two longitudinally split chromo- 

 somes without divergence of their synaptic ends. This form of 

 ring has been reported for the Acrididae only by de Sinety and 

 by Montgomery and their figures are of doubtful character. In 

 the Locustidae, Otte and Vejdovsky figure such rings along with 

 others of the first type and it is probable that they overlooked 

 small indications of divergence, for in my own studies of this 

 family no marked differences from the Acrididae appeared. 



Opposed to such a conception of ring formation is one in which 

 the elevations just described as being constituted of diverging 

 members of two superimposed half rings, are regarded merely 

 as accidental crossings of the free ends of synaptic mates. No 

 reason is assigned for the maintenance of this chance relation 

 through the subsequent chromosome movements during which 

 extensive adjustments take place which might reasonably be 

 supposed to change or obliterate it. Doubtless such relations 

 do occur in the prophase but upon close analysis it becomes very 

 evident that they are not permanent and do not have a meaning 

 opposed to that of the first interpretation. It is most significant 

 that in all the figures drawn by those who hold to the second 

 type not a single one represents a ring en face — the point of view 

 from which the constitution of the ring is indubitable. I think it 



